It appeared upon the trial that Niles, the owner of the barn alleged in the indictment to have been wilfully set fire to and burned by the defendants, kept a watchdog, which usually slept in the kitchen in his house; that the dog was there on the night when the barn was burnt, and did not bark until strangers began to collect around the building, after the alarm of fire had been given. The attorney for the Commonwealth thereupon proposed to introduce evidence in relation to the usual habits of the dog, that he always barked when strangers approached or passed by the house, and that he was silent when inmates of the family, or persons with whom he was familiar, did so. This was objected to by the defendants; and the presiding judge ruled, that evidence in relation to the habits of the dog was inadmissible; but that witnesses might be permitted to relate what, within their personal knowledge and observation, the dog did, or forbore to do, on such occasions; and they accordingly *204did testify of particular instances and occasions on which they had noticed that when strangers came to or passed by the house he gave instant alarm by barking, but was quiet and still if the persons so approaching were known or familiar to him.
It is unnecessary to consider whether the instinct or habits of an animal of this kind might not legally be availed of as competent evidence in a judicial proceeding, in any supposed or possible contingency, for the purpose of showing the presence of an accused party in a particular place, or of establishing his personal identity. The present inquiry in relation to the competency of the evidence objected to is limited to the circumstances under which it was offered and received; and the question of its admissibility must be determined in view of the particular posture and aspect of the case, of the facts which had already been, and the additional facts proposed to be, proved at the time when the objection was taken and urged against it.
It is obvious from the statements in the bill of exceptions, that the government intended to rely, as one circumstance tending to show the guilty connection of the defendants with the criminal act charged against them, upon the supposed fact that the persons by whom it was perpetrated passed by the house of Niles on their way to the bam in which the fire was kindled. The evidence objected to could not have been offered for the purpose of showing that any persons actually did pass along in that direction, for it had no such tendency, but merely to identify the defendants as the persons otherwise shown to have been there. If therefore there were no persons in that place, or, what is the same thing, if there was no evidence, direct or indirect, in the case to indicate or establish that fact, that which was offered to prove the identity of particular individuals, supposed or alleged to be there, was wholly immaterial to the issue to be tried, and should for that reason have been rejected as irrelevant. 1 Stark. Ev. 387.
The silence of the dog until after the people in the neighborhood began to assemble, when the building was discovered to be on fire, is quite as consistent with the fact that no person approached or passed by the house of Niles as with the suppo*205eition that either of the defendants was in its immediate vicinity. To make this silence of the animal therefore a circumstance of any possible importance, or to impute to it any conceivable significance in relation to the designation or identity of either of the defendants, there should first have been proof, or at least an offer or a suggestion of an intention to offer some evidence tending to prove, that some person or persons came or approached so near to the house as then to have been likely to attract his attention. But nothing of this kind appears to have been shown, or attempted or proffered to be shown. The barn was at a distance of eight rods from the house; and it has not been contended that persons might not have been there without having attracted the notice or been detected by the vigilance of the dog. And upon careful scrutiny of all the statements in the bill of exceptions, it is not found, nor was it even suggested in the argument of the counsel for the government, that any evidence, either direct or circumstantial, was given, or proposed to be introduced, upon the trial, designed or having any tendency to show that even a single person approached near to, or passed by, the house previous to the assembling of the people, after the breaking out and discovery of the fire. On the contrary, it was expressly conceded that this, if the most convenient, was not the only practicable way of access to the barn from other parts of the town; but that persons, desiring to keep themselves concealed, or to go where they would be sure not to be noticed by the dog, might reach it without difficulty in boats across the water near to which it was situate. Under such circumstances, certainly the quiet and silence of the dog was utterly without significance in relation to the issue to be tried. Evidence of his conduct or behavior at other times could not be legitimately resorted to for the purpose of showing that his conduct on this occasion tended to identify a particular person, or to point him out and distinguish him from others; because it was in no way shown, or offered to be shown, that any individual was in a situation where and to whom it could possibly be applied. It was therefore entirely irrelevant; and it should for this reason have been excluded.
*206This however was not the precise ground upon which the evidence objected to was offered by the district attorney. The special and definite proposition, upon which its competency was placed and insisted upon by him, and upon which it appears to have been affirmed by the court, is that it was admissible in consequence or by reason of its connection with testimony after-wards to be produced, tending to show that Ross, one of the defendants, in a conversation with a witness, who was in attendance and to be examined upon the trial, had declared their intention to set fire to the barn, saying at the same time that they should not do it until after they left the place, but must do it soon afterwards and before the dog forgot them, so that, if he should see them, he would not bark. But the admissibility of the evidence cannot be sustained or justified upon this ground, because no connection whatever can be traced, and there is in fact no relation to each other, between the two separate and distinct propositions which the government thus proposed to prove. It might be a sufficient answer to the position thus taken on the part of the government, to say that the contingency spoken of by Ross never occurred; for it cannot be pretended that the dog, while he was sleeping or confined in the kitchen, saw or could possibly have seen him or any other person then passing by the house. But it is a more important, and indeed decisive consideration, that these two subjects, of which evidence having been given or offered in relation to one is claimed to lay a sufficient foundation for the introduction of proof concerning the other, were distinct and independent matters, having no connection with or relation to each other. The alleged conversation of Ross, in which he spoke or uttered threats of setting fire to the barn, having occurred while he was still a resident in the house and the occupant of part of the farm of Niles, certainly could have had no influence upon or connection with the acts of the dog on the various other occasions to which the testimony of the witnesses referred. Neither his barking nor his silence in the particular instances of which the witnesses were permitted to testify could have had any tendency to corroborate or confirm the testimony of the other witness, who professed *207to reveal and give an account of what Ross had said concerning the design and intention of Marshall and himself relative to burning of the barn. Whether he uttered the threat or used the language imputed to him could in no degree be inferred, either from the general habits of the dog, or his particular conduct when strangers or members of the family came to the house. In truth, he appears to have growled, or been fawning or quiet, not at all in consequence of anything said about him by Ross, or by anybody else, but just as he happened to be excited by the approach of strangers, or unmoved by the arrival of persons whom he knew. It is certainly safe to assume, at least, that his conduct could not have been influenced or affected by any sayings or declarations made in his absence, of which he never heard, and of which, if he had heard, he could have had no understanding. Proof therefore of his usual habits, or of what he actually did on particular and special occasions, could afford no presumption, or be the basis of any probability, that what had been or should be testified concerning the conversation and supposed threats of Ross was true. The two things, namely, the conversation of the defendant at one time, and the acts and conduct of the dog at another, are wholly unlike, distinct and independent of each other; and because there was no just or adequate reason for assuming that they were so connected or related as to make it necessary or proper that they should both be taken and considered together, the admission, upon that ground, of the evidence objected to was incorrect and erroneous.
It is worthy of observation also, that in regulating the course of the trial in reference to the examination of witnesses the rule prescribed by the court cannot be considered as having been adhered to. It was ruled, and we think correctly, that no evidence concerning the habits of the dog was admissible; and yet the witnesses were allowed to testify what they had severally known him do on the various occasions which had fallen under their observation. The necessary effect of this testimony was to develop and prove the very habits concerning which the court had distinctly ruled that all evidence was inadmissible. An irregularity of this kind could hardly fail to produce some *208confusion in the minds of the jurors, and may have been the occasion of serious prejudice and disadvantage to the defendants.
For these reasons we are of opinion that the evidence objected to was erroneously admitted, and that the exceptions must be sustained, and a New trial granted.